w/ARP auto tracing and refinement

Relevant web site:

Data preparation

Combine high and low resolution data into one *mtz file if necessary

The following is useful, e.g., when higher resolution data is available than was used in the phasing (e.g., SOLVE run).  In this case, there is one set of data (e.g., the solve.mtz output file) with FP, SIGFP, PHIB and a second set (e.g., from Denzo/Scalepack) that was scaled/merged/processed to full resolution (see SOLVE section).  First, the Denzo/Scalepack file must be converted to *mtz format, then the two sets of files must be combined with CAD to generate a single *mtz file.

Run uniqueify

Either the direct output from SOLVE (solve.mtz) or the combined file described in the previous section can be used for a w/ARP run.  It is suggested to then use "uniqueify" to add a freeR column and make all reflections unique:
< 39 lion /SOLVE/WARP> which uniqueify
/keck/CCP4_V40/etc/uniqueify

< 40 lion /SOLVE/WARP> uniqueify
Usage: uniqueify [-s] [-f <label> | -p <fraction>] <input file>[.mtz] [<output file>]

 From the MTZ file which is the first file argument, create another with
 unique reflections and a column of free R flags.  Output is to file named
 by the second file argument, else input with `-unique' appended to its
 basename. Systematic absences are omitted from the output file unless
 the -s switch is specified. If free R flags are already present in
 the input file, use the -f <label> switch. If no free R flags are present
 then the fraction of reflections to be flagged with each indicator
 can be specified with the -p <fraction> switch (default 0.05).

After running uniquiefy, the datafile can be examined again using mtzdump as above.

Run arp_warp_setup.sh

< 53 lion /SOLVE/WARP> which arp_warp_setup.sh
/keck/arp_warp_5.1/bin/IRIX64/arp_warp_setup.sh

< 54 lion /SOLVE/WARP> arp_warp_setup.sh
===========================================================================
|          This is the setup procedure for ARP/wARP version 5.1           |
===========================================================================
This setup must be run for all modes of operation.

 Enter the name of the mtz file: solve_un.mtz
 ----------------------------------------
 Here are the data included at that file.

 OVERALL FILE STATISTICS for resolution range   0.000 -   0.250
 =======================
 

 Col Sort    Min    Max    Num      %     Mean     Mean   Resolution   Type Column
 num order               Missing complete          abs.   Low    High       label

   1 ASC    -17      16      0  100.00     -1.7      6.5  52.47   2.00   H  H
   2 NONE     0      35      0  100.00     13.1     13.1  52.47   2.00   H  K
   3 NONE     0      27      0  100.00     10.3     10.3  52.47   2.00   H  L
   4 NONE    0.0   638.9   797   95.42   153.62   153.62  18.23   2.00   F  FP
   5 NONE    0.6    30.2   797   95.42     3.67     3.67  18.23   2.00   Q  SIGFP
   6 NONE    0.0   360.0   797   95.42   178.38   178.38  18.23   2.00   P  PHIB
   7 NONE  0.001   1.000   797   95.42    0.849    0.849  18.23   2.00   W  FOM
   8 NONE -814.6   888.6   797   95.42     0.15     8.78  18.23   2.00   A  HLA
   9 NONE -734.1  1073.4   797   95.42     0.18     8.89  18.23   2.00   A  HLB
  10 NONE -541.0   429.1   797   95.42     0.04     2.82  18.23   2.00   A  HLC
  11 NONE -430.3   371.7   797   95.42     0.04     2.74  18.23   2.00   A  HLD
  12 NONE    0.0    19.0     0  100.00     9.50     9.50  52.47   2.00   I  FreeR_flag
 

 No. of reflections used in FILE STATISTICS    17396
 ----------------------------------------
 

 You need to specify some labels from above for use with wARP

  Native data amplitude: FP
  Native data sigma amplitude: SIGFP

  Now enter the size of the protein in RESIDUES / AU: 220

  Protein size estimated at about 1694 atoms
  Average B factor from Wilson Plot estimated to be 14

 Do you plan to use experimental phases as input (i.e. for mode warpNtrace or warp) (Y/N) ? Y
  Amplitude (weighted) for initial map calculation: FP
  Phase for initial map calculation: PHIB
  Weight (ie FOM). Press <Enter> if amplitude is already weighted : FOM

  How many total cycles of arp/warp you plan to run?  100

  How many refinement cycles between rebuilding (for warpNtrace only) ?  10
  How many molecules per asymmetric unit (for warpNtrace only)?  1

  For warpNtrace and molrep modes (restrained modes)
  a proper weight must be set for Xray/geometry contributions.
   Matrix suggested Default: 0.5
   -Decrease to tighten geometry
   -Increase to increase X-ray terms contribution
  Enter an appropriate number (Enter for default)

 Do you plan to use free atom model density modification (mode warp) (Y/N) ? Y

  wARP usually makes 3 big iterations for refining free atom models.
  Typical values for each cycle is 10-30, read the manual for details.
  Note that for good starting maps you can skip the  2nd and 3rd round.
  (just specify 0 cycles for this round)
   How many refinement cycles for 1st wARP iteration ? 15
   How many refinement cycles for 2nd wARP iteration ? 15
   How many refinement cycles for 3rd wARP iteration ? 15

 Do you want multiple free atom models averaging (Y/N) ? Y

  How many models do you plan to use for averaging ? 4

   You will be now asked how many processors you can use at the SAME time
   for running arp/warp jobs. Remember that these machines should share
   a common home directory.

  If you are not sure of what you are doing please consult the local System manager.

  How many processors can you use simoultaneously ? 2
    Processor 1 is in a machine named: lion
     Machine lion is OK.
    Processor 2 is in a machine named: lion
     Machine lion is OK.

 Multiple models setup finished.

  You can choose between the following REFMAC protocols:
    F   A fast protocol that works with good data.
    S   A considerably slower one which might work better in difficult cases.
    R   The slow protocol together with Rfree.
    P   Phased maximum likelihood refibement.
    O   The good old SFALL ...
    H   Optimised parameters for starting from heavy atoms alone.
    W   Optimised parameters for solvent building.
    A   Advanced mode for setting parameters manually.
  What is your choice  ? (F/S/R/P/O/H/W/A) F

============================================

 Succesfull termination.
 See warp.par for the automatic parameters setup

The result from the above setup is an output file which sets parameters for the run:

Run arp_warp.sh mode warpNtrace

side_dock.sh
First, make sure the protein sequence is in PIR format:
drrd.pir (any organization of single letters OK; blank line betweein header and sequence).

The command includes a name for sequence file (no extension) and the desired segment id's (A,B,C...) to be docked.
The segment id's are from the final warpNtrace.brk file that contains VAL, SER. ALA, and GLY.

Example command line format:

side_dock.sh drrd chains A B C D E
Output file:  db_side_dock.log
 

Example trace images

Example maps